Re: [boost] Clang: Open-source C/C++ front end under development

From: David A. Greene (greened_at_[hidden])
Date: 2007-09-02 22:47:36

Bo Persson wrote:
> :: &a[0];
> ::
> :: The C++ community promotes std::vector as a C array replacement,
> :: but
> :: it's really not because it's not legal to do the above operation
> :: when
> :: the vector is empty.
>
> That's just a misuse of the std::vector. You never, ever get the idea 
> of taking the address of an empty array. Why attempt that on an empty 
> vector??
>   

Dunno what you mean by "get the idea of taking the address," but this is
a perfectly
reasonable thing to do in C.  In llvm, there are many interfaces that
want, for
example:

foo(int *array, int num)

It's perfectly legal to pass (&a[0], 0) to that.  It is not legal if a
is a std::vector, however.

That's why std::vector is not a replacement for a C array.  Yet it is
touted as such by C++
proponents.  We need to be much more precise in our advocacy.

> :: I believe this is because the vector
> :: interface
> :: is characterized in terms of iterators rather than addresses.
>
> Yes, so a.begin() works even for an empty vector. Why is the address 
> so interesting?
>   

Because it's the whole point of my message.  The pointer interfaces do
cause some problems.
I'm slowly converting these to take iterators instead.  I have to
respectfully disagree with Chris.
The problem isn't iterators _per_se_, it's the pointer interface the
iterators are used with that
causes the problems.  Converting to these interfaces to iterators in
turn drives a conversion to
use templated  interfaces and this is something the llvm community has
been open to, to its
credit.

                                                  -Dave